Quantitative Analysis of Open Source Projects on SourceForge

Relatively easy accessibility of high volumes of information about open source software makes it an interesting target for quantitative analysis meant to discover some hidden properties and trends of this software development model. In this work we demonstrate how such information can be acquired from the largest open source hosting facility — SourceForge — with nearly minimal effort. We compare our data with an identical data set collected a few months earlier by the OssMole [2] project, of which we were not aware at the time of performing the experiment, but which allowed us to make some interesting cross-comparisons and derive conclusions about temporal changes going on at SourceForge.